The first plot from ggplot in a list is always wrong

Question

I want to save multiple plots to a pdf file. So I tried to use a list to save the results of ggplot. But the problem is the first place of the list is always wrong with that error:

Error: Aesthetics must be either length 1 or the same as the data (651): x, y, colour.

And the rest of the list doesn't have this problem. The code is here:

raw_data <- read.csv("hourly_42401_2018.csv")
raw_data <- separate(data=raw_data,col=Date.Local,into=c("year","month","day"),sep="-")
data <- raw_data[raw_data$State.Name=="Alabama"&raw_data$County.Name=="Jefferson"&raw_data$Site.Num=="23",c("Time.Local","Sample.Measurement","month","day")]
data$Time.Local <- as.numeric(data$Time.Local)-1

m <- c("Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec")
num_m <- unique(data$month)
filename=paste("D:/test/temp1","/","SO2",".pdf",sep="")

plot_list <- list()
for(i in 1:length(num_m))
{
  data_m <- data[data$month==num_m[i],]
  if(nrow(data_m)==0)
    next

  title <- paste("SO2",",",m[as.numeric(num_m[i])],",",2018,",","Alabama",",","Jefferson",",","23",sep="")

  plot_list[[i]] <- ggplot(data_m,aes(x=data_m$Time.Local,y=data_m$Sample.Measurement,colour=data_m$day))+
      geom_point()+geom_line()+xlab("Time")+ylab("parts per million")+labs(fill="Day")+ggtitle(title)+
      theme(plot.title = element_text(hjust = 0.5))+scale_x_continuous(breaks=seq(0,24,6))
  #ggsave(filename,width=50,height=20,units="cm",dpi=300)
}
pdf(filename)
for(i in 1:length(num_m))
{
  p <- plot_list[[i]]
  print(p)
}
dev.off()

And the list description is here: enter image description here Thank very much! It's my first time to ask for help on this website. I am sorry I forgot the sample input and result I expected.The raw data is here:https://drive.google.com/file/d/1Ny3i_W0H9-w0WjXTPZtdQ6QyKXzdv9Va/view?usp=sharing. The data frame data_m is the input data:

 > head(data_m)
    Time.Local Sample.Measurement month day
652          0                0.4    02  01
653          1                0.3    02  01
654          2                0.2    02  01
655          3                0.2    02  01
656          4                0.1    02  01
657          5                0.2    02  01

And the result I am looking for is a pdf file with plots on each page.One of the plots is like that: enter image description here

When asking for help, you should include a simple [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output that can be used to test and verify possible solutions. — MrFlick, Jul 30 '18 at 18:42
I cannot reproduce the error with the sample data you provided. If we cannot reproduce the problem, it will be very difficult to attempt to help you. — MrFlick, Jul 30 '18 at 19:51
I am sorry for my mistake. I have already upload the raw data. Thanks for all you guys' help! — Jason Yang, Jul 30 '18 at 20:30

Jean-Rémi Telmon · Answer 1 · 2018-07-30T22:44:36.377

As mentioned you should include someway to recreate your data

Anyway this should work

    library(tidyverse)

m <- c("Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec")
raw_data <- read_csv("hourly_42401_2018.csv")

aw_data <- separate(data=raw_data,col="Date Local",into=c("year","month","day"),sep="-")
data <- aw_data %>% 
  filter(`State Name`=="Alabama"&&
         `County Name`=="Jefferson"&&
         `Site Num`=="0023")


num_m <- unique(data$month)

plot_list <- list()
for(i in 1:length(num_m)){
  data_m <- data[data$month==num_m[1],]

  if(nrow(data_m)!=0){
    title <- paste(
      "SO2",
      m[as.numeric(num_m[i])],
      2018,
      "Alabama",
      "Jefferson",
      "23",sep=",")

    plot_list[[i]] <- 
      ggplot(
        data_m,
        aes(
          x = `Time Local`,
          y = `Sample Measurement`,
          colour = day
          )
        )+
      geom_point()+
      geom_line()+
      xlab("Time")+
      ylab("parts per million")+
      ggtitle(title)+
      theme(plot.title = element_text(hjust = 0.5))+
      scale_x_continuous(breaks=seq(0,24,6))    
  }

}
pdf("filename.pdf")
for(i in 1:length(num_m)){
  p <- plot_list[[i]]
  print(p)
}
dev.off()

I'm not sure what I change, I retyped everything. Make your to restart your R session, especially with pdf and dev.off

Yep but its better to type dput(data) in the console et copy paste the output so one can recreated your data. Like mine for data gives structure(list(Time.local = c(0, 1, 2, 3, 4, 5), Sample.Measurement = c(0.1, + 0.2, 0.3, 0.4, 0.5, 0), month = c(1, 2, 3, 4, 1, 2), day = c(1, + 5, 3, 6, 5, 8)), class = "data.frame", row.names = c(NA, -6L)) — Jean-Rémi Telmon, Jul 30 '18 at 21:32
Right. I tried your code on my local computer, it worked. But I think the problem is the size of the elements in each position of the list. So I upload all the data. This problem is so weird that when I use your test data, everything is OK. But when I turn to my raw data, the error appears again. — Jason Yang, Jul 30 '18 at 21:46
@JasonYang I downloaded your data and tried. No problem. I edited my code because I used read_csv and it produces tibbles that accept space in cols name. PDF Outputed is 3Mo (and slowdown my computer a lot, but ok) — Jean-Rémi Telmon, Jul 30 '18 at 22:46

score 0 · Answer 2 · answered Jul 30 '18 at 23:42

0

Your problem is that

ggplot(data_m,aes(x=data_m$Time.Local,y=data_m$Sample.Measurement,colour=data_m$day))+

should not include data_m$ inaes(). Your code will not use the provided data for plotting.

answered Jul 30 '18 at 23:42

TC Zhang

2,757
1
13
19

The first plot from ggplot in a list is always wrong

2 Answers2